Quantum 2025 – scientific programme
Parts | Days | Selection | Search | Updates | Downloads | Help
FRI: Friday Contributed Sessions
FRI 7: Entanglement and Complexity: Contributed Session to Symposium III
FRI 7.4: Talk
Friday, September 12, 2025, 11:30–11:45, ZHG008
Demonstration of verified BosonSampling — •Naomi Spier1, Redlef B G Braamhaar1, Riko Schadow2, Sara Marzban1, Jens Eisert2, Nathan Walk2, and Jelmer J Renema1 — 1University of Twente — 2Freie Universität Berlin
Sampling from random quantum circuits has been proposed as a first demonstration of a concrete computational advantage for special purpose, non-universal quantum processors. Whilst ideal sampling implementations have extremely strong complexity theoretic arguments for their classical intractability, real sampling devices are vulnerable to the potential existence of efficient, classical simulation algorithms and the status of many claimed advantage demonstrations remains contested. The original sampling proposal, BosonSampling, involves the propagation of single photon states through a random linear optical interferometer and has received significant attention, especially due to the rapid increase in the size, quality and configurability of integrated photonic waveguides. Whilst photon loss errors are immediately apparent from the data, errors due interferometer imperfections and photon distinguishability can also destroy quantum advantage but are more challenging to quantify. In this work, utilising recently developed photonic fidelity witnesses, we carry out a proof-of-principle, efficient verification of a BosonSampler using an integrated, reconfigurable interferometer. The verification is shown to detect errors due to interferometer noise, distinguishability and Poissonian photon sources.
Keywords: Photonics; Random Sampling; Quantum advantage; Benchmarking; Certification